Author Login Editor-in-Chief Peer Review Editor Work Office Work

Computer Engineering

Previous Articles    

Binary Feature Matching Approach Based on Sequential Sampling Evaluation

HE Li,ZHOU Chuanwei,ZHANG Kun,SHI Chaoxia   

  1. (School of Computer,Nanjing University of Science and Technology,Nanjing 210094,China)
  • Received:2016-08-05 Online:2017-08-15 Published:2017-08-15



  1. (南京理工大学 计算机学院,南京 210094)
  • 作者简介:贺黎(1995—),男,硕士研究生,主研方向为人工智能;周传伟、张坤,硕士研究生;石朝侠,副教授、博士。
  • 基金资助:

Abstract: Aiming at the problem of computating resources waste in sampling process for the classical feature matching algorithm,this paper proposes a new binary feature matching approach.Local features use strong corners as the keypoints and rotation-invariant binary string as the feature descriptor.Sequential sampling evaluation uses Hammingdistance to sort matching pairs,the samples are selected sequentially,the mismatching pairs are eliminated by the model calculated with the least squares method.Experimental results show that the proposed approach can decrease the computation time significantly while achieving the same accuracy compared with the RANSAC and PROSAC algorithms.

Key words: feature matching, Hamming distance, sequential sampling evaluation, least squares method, PROSAC algorithm

摘要: 针对特征匹配经典算法在抽样过程的随机性造成计算资源浪费的问题,提出一种新的二进制特征匹配方法。局部特征以强角点为关键点,采用旋转不变二进制串进行特征描述。顺序采样评估运用Hamming距离对匹配对进行排序,顺序选取样本,利用最小二乘方法拟合的模型剔除误匹配并更新样本和最优解。实验结果表明,与PROSAC及RANSAC算法相比,该方法在保证相同精度的前提下运行时间明显缩短。

关键词: 特征匹配, Hamming距离, 顺序采样评估, 最小二乘法, PROSAC算法

CLC Number: